Reducing Overtreatment in Cancer Patients: When Is Active Surveillance Appropriate?
Michigan Medicine experts discuss whether active surveillance — close monitoring without immediate treatment — could reduce overtreatment for some thyroid, prostate and breast cancer patients.
For certain slow-growing or low-risk cancers, surgery and other treatments might not be the best course of action.
In a new perspective piece in the New England Journal of Medicine, researchers from the University of Michigan Comprehensive Cancer Center examine whether active surveillance could be a strategy to reduce overtreatment in certain patients.
Among the groups best suited for consideration: those with low-risk thyroid cancer, prostate cancer and a form of early breast cancer called ductal carcinoma in situ, or DCIS.
There’s no question that cancer treatment comes with risk: mastectomy complications and lymphedema for women with breast cancer; impotence and incontinence for men with prostate cancer; voice changes and low calcium levels for patients with thyroid cancer. And it can be costly for patients and the health care system.
By comparison, active surveillance involves monitoring patients without engaging in surgery or intensive therapies. It includes regularly following up with patients and performing biopsies and tests when necessary. Patients who show signs of advancing cancer receive immediate treatment.
Still, many physicians and patients are reluctant to opt for active surveillance.
Each of the perspective’s authors research overtreatment within his or her respective specialty. Megan Haymart, M.D., associate professor of internal medicine, treats patients with thyroid cancer. David C. Miller, M.D., professor of urology, sees patients with prostate cancer. Sarah T. Hawley, Ph.D., researches patient and provider preferences around breast cancer treatment.
All three authors spoke more about the subject:
Why are patients with certain cancers better candidates for active surveillance?
Haymart: Active surveillance is not an option for all types of cancer of the breast, prostate and thyroid. But some subgroups of those cancers may qualify for active surveillance because they have disease that is at low risk for progression and cancer-related death. Although some patients with thyroid cancer have aggressive disease, many more have relatively indolent disease. In these patients, the risks of treatments may exceed any perceived benefits.
Miller: The key aspect of prostate cancer is that it is heterogeneous. Many cancers grow slowly, and other cancers grow rapidly. Historically, there may have been a mismatch for some men between the aggressiveness of the cancer and the aggressiveness of the treatment. Many low-risk prostate cancers are not destined to grow and spread. The challenge is that you may give aggressive treatment without benefiting the patients because the cancer was never destined to misbehave.
Hawley: DCIS, also known as stage 0 breast cancer, is completely contained; it is not invasive. The current treatment is usually lumpectomy with radiation therapy, though treatment with lumpectomy only has been proposed. The goal is to identify the very low-risk DCIS that is least likely to develop into invasive breast cancer. This may be where active surveillance could be a viable option.
When should active surveillance be considered?
Haymart: If the risk of treatment is far greater than the risk of cancer-related complications or death, then that would suggest considering active surveillance. For all cancers, there’s a risk from surgery, and some cancer treatments may not have long-term survival benefits. So it gets down to balancing benefits and risks.
Based on data from clinical trials, active surveillance appears to be an option for some cases of low-risk thyroid cancer. However, active surveillance of thyroid cancer has not yet been implemented broadly, and there are several necessary steps prior to its widespread implementation. There should first be a careful strategy about what constitutes appropriate active surveillance, who is the optimal candidate and how to reduce harms to patients, including identifying when cancer progresses and reducing patient worry related to the cancer diagnosis.
Hawley: Of the three low-risk cancers, DCIS may be the most difficult for active surveillance because we don’t have specific markers or clear clinical tests to identify who is the best candidate for active surveillance. Although a stage 0 classification is an indicator, other factors like older age and positive hormone receptor status should be considered in thinking about offering active surveillance.
DCIS has a near 100 percent survival rate. Yet we have been seeing a shift toward double mastectomy to treat DCIS — a much more aggressive surgery that does not lead to improved survival for most of these patients. In considering active surveillance, we need to be sure we understand and support the patient decision-making process and consider how patients can feel satisfied with decisions for less-extensive treatment in the context of low-risk DCIS.
Miller: For prostate cancer, while many selection criteria have been proposed, the best candidates generally have a Gleason score of 6, a PSA value of 10 or less and a clinical T stage of T2a or lower.
How did active surveillance become standard in prostate cancer?
Miller: It has been increasingly recognized that many men with low-risk prostate cancer survive for a long time even without aggressive surgery or radiation treatment. Many pioneering centers have been doing active surveillance for nearly 20 years, and it has become more widely accepted and adopted in the past decade.
There has been increased comfort with the safety of this process and the idea that it really is an active management strategy. Currently, most data indicate that 30 to 50 percent of men with low-risk cancers are entering active surveillance.
Is there evidence of cost savings in active surveillance?
Miller: There have been a number of analyses looking at this question. Generally, in the first few years after diagnosis, the overall costs of active surveillance are lower than the costs associated with definitive treatment. However, if someone is on active surveillance for a long time, perhaps five years or longer, and has been receiving multiple tests, then the cumulative costs may end up exceeding the costs of surgery or radiation.
Still, there is a benefit if we are helping patients avoid side effects of treatment for cancers that are very low risk.
What’s the toughest aspect of getting physicians and patients to choose active surveillance?
Haymart: I think that by and large, physicians want to provide good care for their patients, and they worry about missing the unknown. Uncertainty makes doctors uncomfortable; they feel like doing more equals better care. But this is not always the case. Also, patients and physicians have different levels of comfort with what are acceptable risks. So this is where there’s room for education for physicians and patients.
Miller: A major concern among physicians is the possibility that we are exposing patients to avoidable cancer progression. From both the patient and physician’s perspective, one patient that has this happen is one too many. In addition, patient anxiety about living with a cancer diagnosis is a potential barrier to surveillance. We really have to select the right patients, reassure them about the low-risk nature of the cancer and then follow them carefully over time for any changes that suggest things are becoming more aggressive.
What are some concerns or limitations related to active surveillance?
Haymart: For thyroid cancer, there are several steps that have to take place. We need to identify what optimal active surveillance is and to determine who would be appropriate candidates. Both physicians and patients need to be on board with active surveillance as a management option. Plus, if implemented, we need to make sure that the effect of the cancer on the emotional health of the patient is monitored, because fear and anxiety are often a chief concern.
Miller: There are legitimate concerns about underestimating the risk of the cancer. For example, could there be some higher-grade cancer, Gleason 7 or higher, that is hiding out in the prostate? So an important advance for active surveillance has been strategies to better clarify the amount and nature of cancer in the prostate. Specifically, prostate MRI has emerged in the last several years and, in my opinion, now provides the best imaging window into the prostate. This technology has helped me feel more comfortable about who has low-risk cancer. Also, emerging genomic tests hold the potential to identify truly low-risk patients based on characteristic patterns of mutations.
Hawley: In breast cancer, we don’t have the history of outcomes following active surveillance that is available for prostate cancer. There is a study in the United Kingdom comparing active surveillance to treatment for low-risk DCIS that will provide important information on outcomes. In addition, we have to determine which patients will want to undergo active surveillance. Receiving a diagnosis of cancer makes most patients want to take action to remove the threat. People don’t want uncertainty when it comes to cancer. That’s completely understandable. Helping patients to understand that low-risk DCIS does not require immediate action or rushed decision-making is important.
What improvements can be made to active surveillance?
Miller: In addition to identifying the right patients for this management plan, we also have to determine the best way to implement active surveillance. This has to be a partnership with patients. Follow-up testing, including MRIs and prostate biopsies, have to be done so that changes in the behavior of the cancer can be identified early. We also have to better define the clinical criteria that indicate a need to transition from active surveillance to treatment with surgery, radiation or some other modality.
Hawley: We need data on the outcomes of patients who have chosen active surveillance for low-risk DCIS. I’m not sure we’ll have the amount of data that will feel most comfortable to clinicians and patients. Still, this is a first step: People are talking. The answer may be no, it’s not viable for breast cancer, but we need to have this conversation.