Primary Surgery in Metastatic Breast Cancer

Study [Reference]

Cancer registry

Study period (follow-up)

Patients

Group differences

Overall survival (OS)

Surgery vs. no surgery

Other comments

Nguyen 2012 [13]

British Columbia, Canada

1996–2003, (median, 1.9 y)

LRT, 378 (52%); surgery, 255 (67%); RT, 82 (22%); both, 41 (11%); no LRT, 355

LRT group: younger age, better ECOG status, lower T and N stage, lower grade, less LVI, more ER+, less sites metastases, less visceral metastases

5-year OS: LRT 21%; no LRT 14%

Improved OS with LRT: HR 0.78; 95% CI 0.64–0.94; p = 0.009

Improved OS in LRT group: < 50 y, with higher performance status, ER+, clear surgical margins, single metastasis, 1–4 metastases and bone-only metastasis

Dominici 2011 [14]

NCCN, USA

1997–2007

Surgery, 54 (19%); no surgery, 236

Matched non-surgery patients to surgery patients. Surgery group: more lung metastases, less treatment with trastuzumab

Median survival: surgery, 3.5 y; no surgery, 3.4 y

No difference: HR, 0.94; 95% CI 0.83–1.08; p = 0.38

Ruiterkamp 2009 [2]

South Netherlands

1993–2004

Surgery, 288 (40%); BCS, 85 (30%); Mast, 189; no surgery, 440

Surgery group: younger age, less comorbidity, smaller T size, less sites of metastases, less visceral metastases, more radiotherapy and systemic therapy

Median survival: surgery, 31 m; no surgery, 14 m

Improved OS with surgery: HR, 0.62; 95% CI 0.51–0.76

Percentage of patients with stage IV cancer: 4.6%

Gnerlich 2007 [15]

SEER, USA

1988–2003

Surgery, 4,578 (47%); BCS, 1,844 (40%); Mast, 2,485; no surgery, 5,156

Surgery group: younger age, more likely to be white and married, lower T stage, higher grade, more ER+ and PR+

OS at end of study: surgery, 24%; no surgery, 16%

Improved OS with surgery: adjusted HR, 0.63; 95% CI, 0.60–0.66; p < 0.001

Rapiti 2006 [16]

Geneva, Switzerland

1977–1996

Surgery, 127 (42%); BCS 40 (31%); Mast, 87; no surgery, 173

Surgery group: younger age, more likely private, lower T and N stage, more single site metastasis, less visceral metastases, less chemotherapy

5-year DSS: surgery (margins-), 27%; surgery (margins+), 16%; no surgery, 12%

Improved DSS, surgery with neg. margins vs. no surgery: adjusted HR, 0.6; 95% CI 0.4–1.0; p = 0.049

Improved DSS also in younger, LN-, ER+, nonsymptomatic and nonCNS or visceral metastases. Percentage of patients with stage IV cancer: 6.5%

Khan 2002 [1]

NCDB, USA

1990–1993

Surgery, 9,162 (57%); BCS, 3,513 (38%); Mast, 5,649; no surgery, 6,861

Mastectomy more likely in women with a single metastasis. Higher proportion of women with soft-tissue/bone metastases vs visceral metastases received mastectomy

3-year OS: mastectomy, 32% (+/-margins: 26%/ 36%); BCS, 28% (+/- margins: 26%/35%); no surgery, 17%

Improved OS with surgery (p < 0.0001): negative margins HR, 0.61 (95% CI 0.58–0.65); positive margins HR, 0.75 (95% CI 0.71–0.79)

Improved OS also with a single metastasis, without visceral metastases and with systemic therapy. Percentage of patients with stage IV cancer: 4.1%

BCS, breast-conserving surgery; CNS, central nervous system; CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; DSS, disease-specific survival; ER, estrogen receptor; HR, hazard ratio; LRT, locoregional treatment; LVI, lymphovascular invasion; m, month; Mast, mastectomy; N, node; NCCN, National Comprehensive Cancer Network; NCDB, National Cancer Data Base; OS, overall survival; PR, progesterone receptor; RT, radiotherapy; SEER, Surveillance, Epidemiology, and End Results; T, tumor; y, year.

Table 15.2

Institutional studies of primary surgery in MBC

Study [Reference]	Institution	Study period (follow-up)	Patients	Group differences	Overall survival (OS)	Surgery vs. no surgery	Other comments
Lang 2013 [17] (update of Babiera 2006 [32])	M.D. Anderson Cancer Center, USA	1997–2002 (med. 74 m)	Surgery, 74 (36%); BCS, 33 (45%); Mast, 41; no surgery, 134	Surgery group: lower T and N stage, less sites of metastases, less multiorgan metastases, more received chemotherapy only, more received RT	Median survival: surgery, 56 m; no surgery, 37 m	Improved OS with surgery: HR 0.58; 95% CI 0.35–0.98; p = 0.04	Improved OS also in ER+ and a single site of metastasis. Surgery before stage IV diagnosis: 30 patients (41%)
Rashaan 2012 [19]	Hospitals in Leiden and ’s- Hertogenbosch, Netherlands	1989–2009	Surgery, 59 (35%); BCS, 11 (19%); Mast, 48; no surgery, 112	Surgery group: younger, less medication use, smaller T size, more with a single site of metastasis	NR	No overall difference on multivariate analysis: HR 0.9; 95% CI 0.6–1.4; p = 0.5	Improved OS in younger patients and patients without comorbidity that received surgery. Surgery before stage IV diagnosis: 21 patients (36%)
Samiee 2012 [18]	Ottawa and Queensway Carleton Hospitals, Canada	2005–2007 (med. 40 m)	Surgery, 48 (43%); no surgery, 63	No significant differences, but lower proportion of patients with visceral metastasis and advanced T and N stages in surgery group.	Mean survival: surgery, 49 m; no surgery, 33 m	Improved OS with surgery: p = 0.016	Surgery before stage IV diagnosis: 29 patients (60%)
Roche 2011 [20]	Universitätsmedizin, Berlin, Germany	1986–2007	Surgery, 35 (57%); BCS, 13 (37%); Mast, 22; no surgery, 26	Surgery group: younger, more with single site of metastasis, more received RT	NR	No difference: p = 0.253
Pérez-Fidalgo 2011 [21]	Hospital Clinico Universitario, Valencia, Spain	1982–2005 (med. 30 m)	Surgery, 123 (59%); BCS, 10 (8%); Mast, 113; no surgery, 85	Surgery group: better performance status, more with single site of metastasis, less with visceral metastasis	Median survival: surgery, 40 m; no surgery, 24 m	Improved OS with surgery: HR 0.52; 95% CI 0.35–0.77; p = 0.001	Improved OS also in ER+. Surgery before stage IV diagnosis: 78 patients (63%)
Pathy 2011 [3]	University Malaya Medical Centre, Malaysia	1993–2008	Surgery, 139 (37%); BCS, 6 (4%); Mast, 133; no surgery, 236	Surgery group: less likely to be Malay, lower T and N stage, more with a single site of metastasis, more likely to receive RT and hormone therapy	Median survival: surgery, 21 m; no surgery, 10 m	Improved OS with surgery: adjusted HR 0.72; 95% CI 0.56–0.94)	Improved OS in surgery patients who had negative margins and were < 65 y. Percentage of patients with stage IV cancer: 10.2%
Leung 2010 [23]	Medical College of Virginia, USA	1990–2000	Surgery, 52 (34%); no surgery 105	Surgery group: younger patients and lower N stage	Median survival: surgery, 25 m; no surgery, 13 m	12 m survival increased in surgery group by Wilcoxin test but not by log-rank	Chemotherapy was the only factor associated with improved survival on multivariate analysis
Neuman 2010 [22]	Memorial Sloan- Kettering Cancer Center, USA	2000–2004	Surgery, 69 (37%); BCS, 41 (59%); Mast, 28; no surgery, 117	Surgery group: more likely to have a smaller tumor, be negative for HER2/neu and have a solitary metastasis	Median survival: surgery, 40 m; no surgery, 33 m	No difference on multivariate analysis: HR 0.71; 95% CI 0.47–1.1; p = 0.1	Improved OS in ER+, PR+, HER2/neu amplified and without visceral metastases. Surgery before stage IV diagnosis: 34 patients (49%)
McGuire 2009 [25]	Moffitt Cancer Center, USA	1990–2007 (med. 37 m)	Surgery 154 (27%); BCS, 56 (36%); Mast, 98; no surgery, 412	Surgery group: older patients	OS at median 37 m: surgery, 33%; no surgery, 20%	Improved OS with surgery: p = 0.0012	Improved OS in surgery group seen in mastectomy patients (preoperative chemotherapy more likely)

15.3.1 Study Strengths

One obvious advantage of these studies is the number of patients included in them. The outcomes of more than 27,000 patients and 4,000 patients have been analyzed in the published registry studies and institutional series, respectively. These include patients not only from the United States and Canada, but also from several European countries as well as Malaysia and Japan. They provide valuable epidemiological information, for example, the overall incidence of MBC, which ranges from 4–5% in recent United States and European reports. They also illustrate which treatments patients with MBC are actually receiving. The first and largest report, by Khan et al, was received with surprise; 57% of 16,023 patients with MBC had surgical excision of the primary cancer performed [1]. This percentage was much greater than anticipated. It proved not to be unique; between 40–52% of women in other cancer registry studies, and 27–60% in institutional studies, also had surgical resection of their primary breast cancer. The proportion of patients who underwent breast-conserving surgery (BCS) varied widely among the 16 studies that recorded this information. In 12 of these, between 22–59% of patients had BCS; however, in four studies, the rate of BCS was > 10%.

15.3.2 Study Weaknesses

The published reports on primary surgery in MBC are all retrospective. By their nature, they frequently contain incomplete and uncertain information. This is especially true in the case of tumor registry studies; for example, in the report by Khan et al., detailed histological information was unavailable, the tumor (T) stage was collected instead of the tumor size, data on margins were unavailable for 30% who had surgery, and although external beam radiation therapy was used to treat 36% of patients, it is unknown whether this was administered to the breast or to sites of metastatic cancer [1]. Additionally, the institutional studies are often small, and patients were often recruited over a prolonged period of time. Of the 15 institutional studies presented in this chapter, eight contain fewer than 200 patients, and in a further eight studies, patients were accrued over 15 years or more. The ethos and biases of each institution also determine what treatment patients received and how the data are presented.

15.3.3 Overall Survival: Surgery vs. No Surgery

Two-thirds of the studies reviewed in this chapter show an improvement in overall survival following primary surgery in patients with MBC. This includes four of the six registry studies [1, 2, 13, 15], and 10 of the 15 institutional studies [3, 17, 18, 21, 24–26, 28–30]. Two further studies show equivocal benefit [23] or benefit in a subgroup of patients [16

Only gold members can continue reading. Log In or Register to continue