• © 2006 by American Society of Clinical Oncology

Randomized Trial of Long-Term Follow-Up for Early-Stage Breast Cancer: A Comparison of Family Physician Versus Specialist Care

The article by Grunfeld et al¹ in this issue of the Journal of Clinical Oncology is a landmark article for women with breast cancer and their health care providers. It shows conclusively that the health outcomes for women after primary treatment of breast cancer are the same if they are followed by their family physicians or cancer center specialists. Medical, psychosocial, and all other measured outcomes were the same. There really is no other conclusion that can be drawn.

This is not one study; this relates to conclusions from years of research. Grunfeld et al previously showed that in Great Britain, follow-up by a generalist physician led to the same health outcomes as did follow-up by a specialist surgeon,² with no change in quality of life,³ better patient satisfaction,⁴ and no change in health care expenditures.⁵ Another randomized trial in Great Britain showed that twice as many patients preferred simpler, less frequent follow-up by telephone.⁶ There is no a priori reason to expect that patients in the United States would choose differently, and Loprinzi et al demonstrated that the demand for medically inappropriate testing can be reduced by patient education about the specificity, sensitivity, and usefulness of the available tests.⁷ In fact, this conclusion corresponds with the American Society of Clinical Oncology (ASCO) guidelines for surveillance of breast cancer patients.^8,9

Data suggest that current follow-up procedures could be improved. Lash and Silliman followed 303 stage I or II breast cancer patients aged 55 years or older diagnosed at five Boston hospitals, with patient interviews and medical record abstracts.¹⁰ Over 4 years, of the 303 women, 279 had some surveillance testing; the most common test was a mammogram (average, 3.9 in 4 years). Younger women, those treated at a breast cancer center with a unified patient chart, and women who worked, were more likely to complete surveillance; those 75 to 90 years old were less likely to complete guideline-approved surveillance. Mille et al at Centre Regional Leon Berard showed that implementation of clinical practice guidelines on follow-up of patients with localized breast cancer standardized care and a one-third decrease in cost per patient.¹¹ Earle et al showed that patients followed up by a primary care physician and their oncologist were more likely to receive recommended preventive health services than those followed by their oncologists alone¹²; with more and more survivors each year, this will become increasingly important.

With all that evidence, why are we still uneasy with allocating surveillance care to primary care physicians, without oncology routine follow-up? There are substantial issues that have not yet been addressed.

This study looked at the most important medical events, such as recurrence and new primary cancers. Are there data that show how frequently oncologists or primary care physicians inquire about some of the “softer” but important issues of body image, sexuality, sexual functioning, adaptation back to normal life, cognitive dysfunction, neuropathic pain, bone health, depression, menopause, or the myriad symptoms left after modern treatment? We know these issues are important, and we know we can fix some of them; we just don't know in everyday practice how frequently they are addressed and controlled. A prospective study of a one-page follow-up check list would give the answers.

Do patient outcomes for specific symptoms due to treatment, such as chronic breast pain, sexual function, menopause symptoms, depression, and so on vary between the two groups? The Hospital Anxiety and Depression Scale (HADS) and 36-item short form (SF-36) are good instruments for global function, but are not designed for breast cancer survivor symptoms. Are there data to suggest that any practitioner group is better? We do know that oncologists are not very good at assessing emotional health. For example, in a study of 204 patients and five oncologists, patients reported many more symptoms than their oncologists perceived, and the oncologists had better sensitivity and specificity for physical symptoms than psychosocial. The oncologists had sensitivity rates up to 80% for fatigue, nausea, vomiting, and hair loss, but recognized only 17% of the patients with anxiety and only 6% of those with clinical depression.¹³

Do patient outcomes for important issues such as genetic testing and secondary chemoprevention vary between the two groups? No data are provided. Are there data to suggest that oncologists provide good care in this regard? Data from our institution suggest under-referral of patients with potentially detectable genetic changes associated with breast and colorectal cancer for genetic consultation (personal communication, Joann N. Bodurtha, MD, MPH, October 23, 2005) compared with national recommendations,¹⁴ so there is work to be done.

Are there data on the diffusion of new knowledge, such as adding letrozole after 5 years of primary treatment with tamoxifen,¹⁵ switching from tamoxifen to exemestane after 2 to 3 years of primary treatment,¹⁶ or recalling patients who missed out on trastuzamab during their adjuvant therapy? How will ASCO inform members of such practice-altering events? How will family practice and primary care physicians learn? How can patients keep up to date? How well will physicians perform in getting patients to reduce their dietary fat intake, and to exercise? The data from a 2,400- woman randomized controlled trial showed that recurrence risk was reduced by 25% to 49% with reduction in dietary fat from 51 to 31 g/d.¹⁷ That reduction in risk is more than many standard interventions such as chemotherapy or hormonal manipulation; if confirmed in the full published report, it should change practice, hopefully overnight, not after years of diffusion.

What is the best unbiased way to educate all who follow breast cancer patients in the many issues that are present during survivorship? Not all will read the New England Journal of Medicine or the Journal of Clinical Oncology. Pharmaceutical and industry representatives could expand their call circles to include family physicians and primary care physicians about aromatase inhibitors after or in place of tamoxifen, and BRCA1/2 testing. But how about those issues without a drug fix, such as cognitive dysfunction, neuropathy from taxanes, or sexual dysfunction? What safeguards could be put in place to update those who deliver care in a rapidly changing oncology practice? And should we entrust the process of keeping informed to pharmaceutical representatives?

How will oncology nurses and physicians respond when they see no “well” patients? Will this contribute to burnout and loss of job satisfaction?

What are practical steps that can be taken to improve follow-up of women with breast cancer, regardless of who does it?

1. We can work with patient advocacy groups to educate their members on what medical science can and cannot do, and work with them to evolve the most efficient methods of follow-up.

2. Grunfeld et al included the one-page follow-up guideline given to the family physician group. This can be incorporated into the office visits. Such simple prompts appear to help improve communication about important health issues.¹⁸ The ASCO surveillance group can devise a similar one page form with important issues to be addressed. This can be jointly done with primary care physicians.

3. We should experiment with patient-controlled follow-up methods, such as giving each patient a USB flash drive with a form for the health care professional to complete each time. This could list the most common concerns and practical tips for fixing them, along with allergies, problem list, radiographs, and so on. In the absence of a single electronic medical record this may be a useful compromise.

4. We can follow the guidelines we already have in place. They are simple, straightforward, can be done in a short time,¹⁹ and will lead to equal care at less cost. As Loprinzi et al have demonstrated, people will not want tests such as the CA 27.29 and carcinoembryonic antigen if they are not recommended and do not add value.

We recognize that this is a substantial change in practice and attitudes. The only way to demonstrate to “nonbelievers” that generalists and specialists are equal would be to have individual patient level detail on symptoms such as pain, anemia, brain metastases, or other visceral metastatic complications, and that early intervention made a difference. The rarity of such individual events makes such trials impossible; there are no data suggesting that oncologists are better at recognizing those symptoms than primary care health care professionals, and there are no data that early intervention makes a difference. The evidence from multiple randomized clinical trials suggests that ASCO should begin partnering with our primary care colleagues and our shared patients to better improve the follow-up process.