Prognostic Significance of Further Axillary Dissection in Breast Cancer Patients with Micrometastases & the Number of Micrometastases: a SEER Population-Based Analysis

Figures & data

Figure 1.  Selection process of our defined population.

ALND: Axillary lymph node dissection; ER: Estrogen receptor; PR: Progesterone receptor status.

Figure 2.  Tendency of patients with one to two micrometastases undergoing sentinel lymph node biopsy alone between 2004 and 2013.

SLNB: Sentinel lymph node biopsy.

Table 1. Baseline characteristics between nonaxillary lymph node dissection group and axillary lymph node dissection group.

Groups	Subgroups	Non-ALND (n = 3689)		ALND (n = 1971)		Total (n = 5660)		p-value
		No.	%	No.	%	No.	%
Age								<0.001
	18–50 years	893	24	633	55	1526	27
	51–70 years	2032	55	1082	13	3114	55
	>70 years	764	21	256	32	1020	18
Race								<0.001
	White	3080	84	1561	79	4641	82
	Black	296	8	248	13	544	10
	Others/unknown	313	8	162	8	475	8
Year of diagnosis								<0.001
	2004–2008	1418	38	1241	63	2659	47
	2009–2013	2271	62	730	37	3001	53
Tumor grade								<0.001
	I	952	26	406	21	1358	24
	II	1825	49	896	34	2721	48
	III/IV	912	25	849	45	1581	28
Tumor size								0.009
	≤2 cm	2617	71	1334	68	3951	70
	>2 cm, ≤3 cm	819	22	463	23	1282	23
	>3 cm, ≤5 cm	253	7	174	9	427	7
ER status								<0.001
	Positive	3367	91	1671	85	5038	89
	Negative	322	9	300	15	622	11
PR status								<0.001
	Positive	3055	83	1497	76	4552	80
	Negative	634	17	474	24	1108	20
HER2 status (n = 2400)								0.004
	Positive	153	12	56	8	209	9
	Negative	1786	88	405	92	2191	91
Molecular subtype (n = 2400)								<0.001
	HR^+/HER2-	1686	87	361	78	2047	85
	HR^+/HER2^+	129	7	48	10	177	8
	HR^-/HER2^+	24	1	8	2	32	1
	Triple negative	100	5	44	10	144	6
Positive SLNs								<0.001
	1	3406	92	1607	82	5013	89
	2	283	8	364	18	729	11
Chemotherapy								<0.001
	Yes	1815	49	1372	70	3187	56
	No/unknown	1874	51	599	30	2473	44
Median number of LN removed		2 (1-5)		14 (9-48)

ALND: Axillary lymph node dissection; ER: Estrogen receptor; HER2: Human epidermal receptor 2; LN: Lymph node; PR: Progesterone receptor; SLN: Sentinel lymph node.

Figure 3.  Survival analyses of patients with 1–2 micrometastases according to axillary treatment.

ALND: Axillary lymph node dissection; BCSS: Breast cancer-specific survival.

Table 2. Univariate and multivariate analyses of breast cancer-specific survival in unmatched and matched cohort.

Groups	Subgroups	Unmatched cohort			Matched cohort
		Breast cancer-specific survival			Breast cancer-specific survival
		Log-rank p-value	Cox model p-value	HRs^† (95% CI)	Log-rank p-value	Cox model p-value	HRs^† (95% CI)
Age		<0.001			<0.001
	18–50 years		–	1.00			1.00
	51–70 years		0.869	0.97 (0.69–1.37)		0.603	0.90 (0.60–1.35)
	>70 years		<0.0001	2.72 (1.82–4.08)		<0.001	2.94 (1.76–4.91)
Race		<0.001			0.001
	White		–	1.00			1.00
	Black		0.002	1.73 (1.21–2.47)		0.006	1.83 (1.19–2.83)
	Others/unknown		0.947	0.98 (0.58–1.67)		0.664	0.83 (0.36–1.91)
Tumor size		<0.001			<0.001
	≤2 cm		–	1.00			1.00
	>2 cm, ≤3 cm		<0.0001	2.17 (1.60–2.96)		<0.001	2.33 (1.62–2.36)
	>3 cm, ≤5 cm		<0.0001	3.37 (2.31–4.93)		<0.001	3.55 (2.20–5.72)
Tumor grade		<0.001			<0.001
	I		–	1.00			1.00
	II		0.003	2.57 (1.39–4.75)		0.008	3.15 (1.35–7.39)
	III/IV		<0.0001	4.49 (2.38–8.46)		0.001	4.55 (1.90–10.95)
ER status		<0.001			<0.001
	Negative		–	1.00			1.00
	Positive		0.009	0.58 (0.38–0.87)		0.021	0.55 (0.33–0.91)
PR status		<0.001			<0.001
	Negative		–	1.00			1.00
	Positive		<0.0001	0.47 (0.32–0.68)		<0.001	0.42 (0.26–0.68)
Axillary treatment		0.460			0.460
	Non-ALND		–	1.00			1.00
	ALND		0.783	0.96 (0.73–1.27)		0.725	0.94 (0.68–1.31)
Positive SLNs		0.409			0.409
	1		–	1.00			1.00
	2		0.477	0.86 (0.56–1.32)		0.838	0.94 (0.53–1.67)
Chemotherapy		0.383			0.383
	No/unknown		–	1.00			1.00
	Yes		0.089	0.74 (0.52–1.05)		0.041	0.63 (0.40–0.98)

^†Adjustment for race, age, tumor size and tumor grade.

ER status, PR status, axillary treatment, positive number of SLNs, adjuvant chemotherapy.

ALND: Axillary lymph node dissection; ER: Estrogen receptor; HR: Hazard ratio; PR: Progesterone receptor; SLN: Sentinel lymph node.

Table 3. Patient demographics after propensity score matching. 

Groups	Subgroups	Non-ALND (n = 1703)		ALND (n = 1703)		Total (n = 3406)		p-value
		No.	%	No.	%	No.	%
Age								0.881
	18–50 years	502	30	515	30	1017	30
	51–70 years	969	57	956	56	1925	57
	>70 years	232	14	232	14	464	14
Race								0.970
	White	1412	83	1417	83	2829	83
	Black	180	11	178	11	358	11
	Others/unknown	111	7	108	6	219	6
Tumor size								0.322
	≤2 cm	1161	68	1197	70	2358	69
	>2 cm, ≤3 cm	418	25	381	22	799	24
	>3 cm, ≤5 cm	124	7	125	7	249	7
Tumor grade								0.951
	I	360	21	362	21	722	21
	II	810	48	801	47	1611	47
	III/IV	533	31	540	32	1073	32
ER status								0.639
	Negative	205	12	214	13	419	12
	Positive	1498	88	1489	87	2987	88
PR status								0.214
	Negative	389	23	359	21	748	22
	Positive	1314	77	1344	79	2658	78
Positive SLNs								0.957
	1	1509	89	1510	89	3019	89
	2	194	11	193	11	387	11
Chemotherapy								0.971
	Yes	560	33	561	33	1121	33
	No/unknown	1143	67	1142	67	2285	67
Median number of LNs removed		2 (1–5)		14 (9–43)

A total of 268 patients derived from ALND group were excluded during propensity score matching.

ALND: Axillary lymph node dissection; ER: Estrogen receptor; LN: Lymph nodel; PR: Progesterone receptor; SLN: Sentinel lymph node.

Figure 4.  Subgroup analyses assessing the benefit of further axillary lymph node dissection according to certain characteristics.

ALND: Axillary lymph node dissection; BCSS: Breast cancer-specific survival; ER: Estrogen receptor; PR: Progesterone receptor status.

Table 4. Cox regression model for breast cancer-specific survival in patients with one to three micrometastases.

Number of micrometastases	HR (95% CI)^†	p-value	HR (95% CI)^‡	p-value
1–2 micrometastases (n = 6284)	1.00	0.003	1.00	0.006
3 micrometastases (n = 155)	2.22 (1.31–3.75)	–	2.10 (1.24–3.56)	–

^†Adjustment for administration of chemotherapy.

^‡Adjustment for race, age, tumor size, tumor grade, estrogen receptor status, progesterone receptor status, positive number of sentinel lymph nodes and adjuvant chemotherapy.

BCSS: Breast cancer-specific survival; HR: Hazard ratio

Figure 5.  Breast cancer-specific survival curve for the presence of one to two and three micrometastases.

BCSS: Breast cancer-specific survival.