The recurrence pattern following delayed breast reconstruction after mastectomy for breast cancer suggests a systemic effect of surgery on occult dormant micrometastases

Abstract

The purpose of this study was to characterize the recurrence dynamics in breast cancer patients after delayed reconstruction. We hypothesized that surgical reconstruction might stimulate dormant micrometastases and reduce time to recurrence. All mastectomy breast cancer patients with delayed surgical reconstruction at Haukeland University Hospital, between 1977 and 2007, n = 312, were studied. Our control group consisted of 1341 breast cancer patients without reconstruction. For each case, all patients in the control group with identical T and N stages and age ±2 years were considered. A paired control was randomly selected from this group. 10 years after primary surgery, 39 of the cases had relapsed, compared to 52 of the matched controls. The reconstructed group was analyzed for relapse dynamics after mastectomy; the first peak in relapses was similarly timed, but smaller than for the controls, while the second peak was similar in time and size. Second, the relapse pattern was analyzed with reconstruction as the starting point. A peak in recurrences was found after 18 months, and a lower peak at the 5th-6th year. The height of the peak correlated with the extent of surgery and initial T and N stages. Timing of the peak was not affected, neither was the cumulative effect. The relapse pattern, when time origin is placed both at mastectomy and at reconstruction, is bimodal with a peak position at the same time points, at 2 years and at 5-6 years. The timing of the transition from dormant micrometastases into clinically detectable macrometastases might be explained by an enhancing effect of surgery.

Keywords: Breast cancer; Breast reconstruction; Multivariate regression; Recurrence dynamics; Surgery; Tumor dormancy.

Fig. 1

Inclusion and exclusion criteria employed to achieve case, control, and matched control populations. DCIS ductal carcinoma in situ, LCIS lobular carcinoma in situ, BCT breast-conserving therapy. 1 Recurrence-free follow-up time equal to or longer than the time to reconstruction of the respective matched reconstructed patient

Fig. 2

Recurrence pattern for the reconstructed patients (n = 312) with T = 0 set at reconstruction (red line) and at primary surgery (blue line). X-axis represents time in months. Y-axis represents six-month hazard rate

Fig. 3

Recurrence pattern according to surgical intervention demonstrates an enhanced, but similarly timed, effect by increased extent of trauma. Blue line patients receiving a unilateral implant. Red line patients receiving more extensive surgery. X-axis represents time in months since reconstructive surgery. Y-axis represents six-month hazard rate. DIEP deep inferior epigastric perforator, TRAM transverse rectus abdominis myocutaneous flap

Fig. 4

The probability of recurrence over time, i.e., Crude Cumulative Incidence (CCI), was estimated according to proper nonparametric estimator adjusting for the presence of competing events and was compared between groups by the Gray test. Simple unilateral implant (black line) and more extensive surgery such as DIEP/TRAM or bilateral procedures (red line). X-axis represents time since reconstruction in months. There is no observable difference between the groups

Fig. 5

Subgroup analysis of recurrence pattern by known prognostic factors. Increasing T and N stages are associated with an enhancing effect on preexisting recurrence risk. Left figure demonstrates the recurrence dynamics for node-positive (red line) and node-negative (blue line) reconstructed patients. Right figure demonstrates the recurrence dynamics for reconstructed patients with tumors >2 cm (red line) and ≤2 cm (blue line). X-axis represents months since reconstruction. Y-axis represents six-month hazard rate

Fig. 6

Recurrence pattern for the reconstructed patients and the matched control patients with T = 0 at primary surgery. Red line reconstructed patients. Blue line control patients. X-axis represents months since primary surgery. Y-axis represents six-month hazard rate

Fig. 7

The probability of recurrence over time, i.e., Crude Cumulative Incidence (CCI), was estimated according to proper nonparametric estimator adjusting for the presence of competing events and was compared between groups by the Gray test. Red line reconstructed patients. Blue line matched control patients. Despite matching by age, time of diagnosis, and T and N stage, there is a nonsignificant trend for a more favorable prognosis in the reconstructed patients. X-axis represents time in months since reconstruction/reference day (see m&m) for reconstructed patients and control patients, respectively. Y-axis represents accumulated recurrence-free survival

Fig. 8

Multiple timescale analysis of the hazard ratio for recurrence between the reconstructed patients and the controls in relation to time since reconstruction. Dotted lines represent 95 % CI. X-axis represents months since reconstruction/reference day. Y-axis represents the ratio between the six-month hazard for recurrence between reconstructed patients and controls