Follow-up Imaging Lacking for Many After Breast Cancer Surgery

Miriam E. Tucker

October 27, 2016

Washington, DC — About one third of US women who receive surgical treatment for breast cancer are not receiving appropriate follow-up, new research suggests.

Findings from the National Cancer Database were presented here at the American College of Surgeons Clinical Congress 2016 by surgery resident Taiwo Adesoye, MD, MPH, from the University of Wisconsin, Madison.

Guidelines from the American Cancer Society and the National Comprehensive Cancer Network recommend mammography 6 to 12 months after completion of radiation therapy for breast cancer and annual mammography thereafter, with the goal of early detection of in-breast recurrences or new primary breast tumors.

Yet, the new data analysis revealed that only two thirds of women diagnosed with stage II or III disease receive breast imaging in the first follow-up year after surgery, and nearly two thirds are not receiving annual imaging for the subsequent 4 years. Patient factors, such as low socioeconomic status, combined with tumor- and treatment-specific factors drove the underutilization. But, unlike what has been seen in prior studies, characteristics of the treating institution did not.

Previous observational studies in this area have found rates of follow-up mammography underuse of 55% to 82%, but those studies were restricted to the older Medicare population or single institutions.

"Understanding factors associated with noncompliance is critical to improve guideline adherence," Dr Adesoye said.

In an interview, principal investigator Caprice C. Greenberg, MD, professor of surgery at the University of Wisconsin, told Medscape Medical News, "The thing that's different about this in terms of underutilization is that rates didn't vary based on institution type. Often variation in utilization varies from local hospital to hospital based on local practice patterns. Here, it seems to be pretty consistent across hospital types. Rather, it seems to be this more vulnerable population and patients with more advanced disease."

The message, Dr Greenberg said, is "Don't forget about locoregional recurrences, and think about your vulnerable populations."

Session moderator and discussant Sarah Blair, MD, vice chair of academic affairs, Department of Surgery, and professor of surgery at the University of California, San Diego, told Medscape Medical News, "Patients that are underinsured and poor socioeconomic status may not have continued access to healthcare over time," noting that the population in the study were also at high risk for distant disease and may be undergoing prolonged treatment.

"The patients or their treating physicians may be more concerned with distant spread rather than recurrence in the breast," Dr Blair hypothesized.

The advice, she said, is "to make sure patients have a good survivorship plan when they finish active treatment to make sure if they move or change insurance they know they should be doing mammograms once a year."

Less Than Half Received Follow-up Imaging

Dr Adesoye and colleagues initially analyzed data for over 11,000 women aged 18 years and older who underwent definitive surgery for histologically confirmed stage II or III breast cancer that had been diagnosed during 2006–2007. They restricted the study to women with stage II and II breast cancer because they were trying to maximize recurrence rates.

After exclusion of those who died, developed local or distant recurrence or a new cancer, or had bilateral mastectomy within 10 months of diagnosis, the total study population consisted of 9835 women in the first year It dropped off from there, so that by the fourth study year (48 to 59 months after diagnosis), the study population was 7457 women.

The total group was mostly white, and 60% had no comorbidities. About half had private insurance, while just over a third had Medicare or other government coverage.

Nearly two thirds (59%) had tumors of 2 to 5 cm; 64.5% were node positive; and nearly three quarters each were estrogen receptor (ER)/progesterone receptor (PR) positive, human epidermal growth factor receptor 2 (HER2) negative, and stage II.

About 44% of the women underwent breast-conserving surgery (BCS) with radiotherapy (RT), while about a quarter each had mastectomy with RT or mastectomy alone. Three quarters had chemotherapy.

Most (93%) lived in urban areas; half received care from a comprehensive community cancer center, 29% were treated at a community hospital, and 17% were treated at a teaching/research institution.

New breast events — locoregional or primary — occurred in 5.5%, in a median time of 2.8 years; 12.5% developed distant recurrence, in a median time of 2.5 years.

Receipt of breast imaging declined over the follow-up period, from 66% in the first year to 58% in the fourth year. Only 38% of patients received annual imaging during all 4 years of follow-up, Dr Adesoye reported.

Dr Blair pointed out during the discussion period that the first follow-up mammogram during months 6 to 12 usually reflects postoperative factors, such as residual calcifications and fat necrosis, and that "data from years 2 to 5 is more important."

Patient, Tumor, and Treatment Factors

Compared with patients under 50 years of age, follow-up imaging was more likely for older women up to age 80 years (odds ratios [ORs] by decade ranging from 1.2 to 1.6) but less likely for those aged 80 years and older (OR, 0.89). Overall, the relationship with age was significant (P < .0001).

Black patients were significantly less likely than whites to receive follow-up imaging (OR, 0.79; P = .0001), as were those with Medicaid or Medicare insurance and the uninsured (ORs, 0.81 and 0.78; P = .001).

Compared with women with no comorbidities, those with one and two or more comorbidities were less likely to receive follow-up imaging (ORs, 0.88 and 0.79, respectively; P < .0001).

Tumor characteristics associated with fewer follow-up mammograms included tumors larger than 2 cm (ORs, 0.85 for those 2 to 5 cm and 0.69 for those >5 cm; P < .0001) and nodal status (OR, 0.84 for positive vs negative; P = .0001).

Both ER/PR and HER2 positivity predicted greater likelihood of follow-up imaging, while patients with borderline or unknown ER/PR status were less likely to receive follow-up. Similarly, those with HER2-positive tumors were more likely to be followed compared with those who had HER2-negative tumors, while those with borderline or unknown status were much less likely.

By treatment characteristics, women who had received BCS with RT were much more likely to be followed with imaging than those treated with BCS alone or mastectomy with RT or mastectomy alone. Women who had received chemotherapy were also more likely to receive follow-up imaging (all P < .0001).

However, institutional characteristics had no impact (rural vs urban, P = .881; community vs comprehensive cancer center or teaching/research center, P = .89).

Dr Greenberg cautioned that follow-up rates may differ for women with stage I disease or those with bilateral mastectomy — both of whom were excluded from this study. But for women with unilateral mastectomy, she said, "I worry that people may forget to think about the contralateral breast."

Similarly, she noted, follow-up may be neglected in women with multiple comorbidities because "they have a lot of other stuff going on, so in a 10- to 15-minute visit, you forget about all the preventive stuff. I think that may be what's happening here. We can't really comment on the etiology based on this study, but I think it would be an important thing to dig into."

Dr Adesoye, Dr Greenberg, and Dr Blair have disclosed no relevant financial relationships.

American College of Surgeons (ACS) Clinical Congress 2016. Presented October 19, 2016.

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.