Optimal Treatment Strategies for Resectable Locally Advanced Esophageal Squamous Cell Carcinoma: A Real-World Triple Cohort Analysis Using Propensity Score Matching

Figures & data

Table 1 Comparison of Baseline Clinical Characteristics Among Three Groups Before and After Propensity Score Matching

Clinical Characteristics	Before Matching				After Matching
	NCT (n=188)	NCRT (n=197)	S+CRT (n=330)	P value	NCT (n=80)	NCRT (n=80)	S+CRT (n=80)	P value
Sex^†
Male	152 (80.9%)	173 (87.8%)	280 (84.8%)	0.164	66 (82.5%)	67 (83.8%)	71 (88.8%)	0.503
Female	36 (19.1%)	24 (12.2%)	50 (15.2%)		14 (17.5%)	13 (16.2%)	9 (11.2%)
Age (year)^†
<60	75 (39.9%)	86 (43.7%)	108 (32.7%)	0.033	29 (36.2%)	33 (41.2%)	32 (40.0%)	0.797
≥60	113 (60.1%)	111 (56.3%)	222 (67.3%)		51 (63.7%)	47 (58.8%)	48 (60.0%)
History of smoking^†
No	81 (43.1%)	79 (40.1%)	136 (41.2%)	0.835	33 (41.2%)	40 (50.0%)	29 (36.2%)	0.205
Yes	107 (56.9%)	118 (59.9%)	194 (58.8%)		47 (58.8%)	40 (50.0%)	51 (63.7%)
History of alcohol^†
No	94 (50.0%)	82 (41.6%)	167 (50.6%)	0.110	35 (43.8%)	44 (55.0%)	40 (50.0%)	0.362
Yes	94 (50.0%)	115 (58.4%)	163 (49.4%)		45 (56.2%)	36 (45.0%)	40 (50.0%)
Comorbidity^†
No	126 (67.0%)	128 (65.0%)	227 (68.8%)	0.663	51 (63.7%)	50 (62.5%)	54 (67.5%)	0.789
Yes	62 (33.0%)	69 (35.0%)	103 (31.2%)		29 (36.2%)	30 (37.5%)	26 (32.5%)
ECOG^†
0	140 (74.5%)	149 (75.6%)	249 (75.5%)	0.959	57 (71.2%)	60 (75.0%)	60 (75.0%)	0.824
1	48 (25.5%)	48 (24.4%)	81 (24.5%)		23 (28.7%)	20 (25.0%)	20 (25.0%)
Family history^†
No	174 (92.6%)	183 (92.9%)	303 (91.8%)	0.895	73 (91.2%)	72 (90.0%)	74 (92.5%)	0.855
Yes	14 (7.4%)	14 (7.1%)	27 (8.2%)		7 (8.8%)	8 (10.0%)	6 (7.5%)
Tumor localization^†
Lower	67 (35.6%)	102 (51.8%)	130 (39.4%)	0.002	35 (43.8%)	37 (46.2%)	33 (41.2%)	0.353
Middle	96 (51.1%)	79 (40.1%)	177 (53.6%)		33 (41.2%)	34 (42.5%)	42 (52.5%)
Upper	25 (13.3%)	16 (8.1%)	23 (7.0%)		12 (15.0%)	9 (11.2%)	5 (6.2%)
Clinical T stage^†
T1	2 (1.1%)	1 (0.5%)	4 (1.2%)	<0.001	1 (1.2%)	1 (1.2%)	0 (0.0%)	0.904
T2	28 (14.9%)	13 (6.6%)	71 (21.5%)		11 (13.8%)	9 (11.2%)	10 (12.5%)
T3	140 (74.5%)	175 (88.8%)	244 (73.9%)		63 (78.8%)	63 (78.8%)	66 (82.5%)
T4	18 (9.6%)	8 (4.1%)	11 (3.3%)		5 (6.2%)	7 (8.8%)	4 (5.0%)
Clinical N stage^†
N0^‡	43 (22.9%)	58 (29.4%)	97 (29.4%)	0.228	25 (31.2%)	23 (28.7%)	20 (25.0%)	0.677
N+^‡	145 (77.1%)	139 (70.6%)	233 (70.6%)		55 (68.8%)	57 (71.2%)	60 (75.0%)
Chemotherapy regimens
Taxane-platinum	167 (88.8%)	163 (82.7%)	226 (68.5%)	<0.001	73 (91.2%)	67 (83.8%)	54 (67.5%)	<0.001
PF	10 (5.3%)	28 (14.2%)	57 (17.3%)		2 (2.5%)	9 (11.2%)	16 (20.0%)
No^§	0 (0.0%)	0 (0.0%)	42 (12.7%)		0 (0.0%)	0 (0.0%)	10 (12.5%)
Others^¶	11 (5.9%)	6 (3.0%)	5 (1.5%)		5 (6.2%)	4 (5.0%)	0 (0.0%)

Notes: ^†Variables used for propensity score matching. ^‡N0, no lymph node metastasis, N+, lymph node metastasis. ^§Patient did not undergo chemotherapy. ^¶Albumin paclitaxel alone or S-1 alone.

Abbreviations: NCRT, neoadjuvant chemoradiotherapy; NCT, neoadjuvant chemotherapy; S+CRT, surgery followed by chemoradiotherapy taxane-platinum, platinum drugs/docetaxel/paclitaxel/albumin paclitaxel; PF, platinum drugs/fluorouracil/S-1.

Table 2 Comparison of Surgical Procedures Among Three Groups Before and After Propensity Score Matching

Clinicopathologic Characteristics	Before Matching				After Matching
	NCT (n=188)	NCRT (n=197)	Surgery (n=330)	P value	NCT (n=80)	NCRT (n=80)	Surgery (n=80)	P value
Surgical approach
OE	139 (73.9%)	98 (49.7%)	237 (71.8%)	<0.001	59 (73.8%)	41 (51.2%)	60 (75.0%)	0.011
MIE	45 (23.9%)	95 (48.2%)	84 (25.5%)		19 (23.8%)	36 (45.0%)	19 (23.8%)
Conversion to OE	4 (2.1%)	4 (2.0%)	9 (2.7%)		2 (2.5%)	3 (3.8%)	1 (1.2%)
Surgical types
Sweet	105 (55.9%)	47 (23.9%)	220 (66.7%)	<0.001	49 (61.3%)	23 (28.7%)	52 (65.0%)	<0.001
McKeown	78 (41.5%)	147 (74.6%)	103 (31.2%)		30 (37.5%)	56 (70.0%)	27 (33.8%)
Others	5 (2.7%)	3 (1.5%)	7 (2.1%)		1 (1.2%)	1 (1.2%)	1 (1.2%)
Lymph node dissection counts(n)
<15	38 (20.2%)	71 (36.0%)	53 (16.1%)	<0.001	20 (25.0%)	29 (36.2%)	8 (10.0%)	<0.001
≥15	150 (79.8%)	126 (64.0%)	277 (83.9%)		60 (75.0%)	51 (63.7%)	72 (90.0%)

Abbreviations: NCRT, neoadjuvant chemoradiotherapy; NCT, neoadjuvant chemotherapy; S+CRT, surgery followed by chemoradiotherapy; MIE, minimally invasive esophagectomy; OE, open esophagectomy.

Table 3 Comparison of Surgery-Related Complication Among Three Groups Before and After Propensity Score Matching

Complication	Before Matching				After Matching
	NCT (n=188)	NCRT (n=197)	Surgery (n=330)	P value	NCT (n=80)	NCRT (n=80)	Surgery (n=80)	P value
Postoperative complications
No	144 (76.6%)	146 (74.1%)	278 (84.2%)	0.011	65 (81.2%)	58 (72.5%)	71 (88.8%)	0.033
Yes	44 (23.4%)	51 (25.9%)	52 (15.8%)		15 (18.8%)	22 (27.5%)	9 (11.2%)
Anastomotic leakage
No	167 (88.8%)	165 (83.8%)	310 (93.9%)	0.001	72 (90.0%)	66 (82.5%)	76 (95.0%)	0.038
Yes	21 (11.2%)	32 (16.2%)	20 (6.1%)		8 (10.0%)	14 (17.5%)	4 (5.0%)
Anastomotic Stenosis
No	184 (97.9%)	194 (98.5%)	320 (97.0%)	0.528	80 (100.0%)	79 (98.8%)	79 (98.8%)	0.604
Yes	4 (2.1%)	3 (1.5%)	10 (3.0%)		0 (0.0%)	1 (1.2%)	1 (1.2%)
Pulmonary complications
No	183 (97.3%)	191 (97.0%)	323 (97.9%)	0.798	77 (96.2%)	77 (96.2%)	79 (98.8%)	0.555
Yes	5 (2.7%)	6 (3.0%)	7 (2.1%)		3 (3.8%)	3 (3.8%)	1 (1.2%)
Cardiac complications
No	187 (99.5%)	193 (98.0%)	327 (99.1%)	0.333	80 (100.0%)	79 (98.8%)	80 (100.0%)	0.366
Yes	1 (0.5%)	4 (2.0%)	3 (0.9%)		0 (0.0%)	1 (1.2%)	0 (0.0%)
Wound infection
No	185 (98.4%)	194 (98.5%)	327 (99.1%)	0.738	79 (98.8%)	79 (98.8%)	79 (98.8%)	1.000
Yes	3 (1.6%)	3 (1.5%)	3 (0.9%)		1 (1.2%)	1 (1.2%)	1 (1.2%)
Hoarseness
No	187 (99.5%)	195 (99.0%)	326 (98.8%)	0.75	80 (100.0%)	78 (97.5%)	79 (98.8%)	0.363
Yes	1 (0.5%)	2 (1.0%)	4 (1.2%)		0 (0.0%)	2 (2.5%)	1 (1.2%)
Others
No	181 (96.3%)	193 (98.0%)	323 (97.9%)	0.469	79 (98.8%)	77 (96.2%)	78 (97.5%)	0.599
Yes	7 (3.7%)	4 (2.0%)	7 (2.1%)		1 (1.2%)	3 (3.8%)	2 (2.5%)

Abbreviations: NCRT, neoadjuvant chemoradiotherapy; NCT, neoadjuvant chemotherapy; S+CRT, surgery followed by chemoradiotherapy.

Table 4 Comparison of Pathological Outcomes Among Three Groups Before and After Propensity Score Matching

Pathological Outcomes	Before Matching				After Matching
	NCT (n=188)	NCRT (n=197)	Surgery (n=330)	P value	NCT (n=80)	NCRT (n=80)	Surgery (n=80)	P value
R0 resection
N0	5 (2.7%)	4 (2.0%)	9 (2.7%)	0.876	2 (2.5%)	1 (1.3%)	3 (3.8%)	0.599
Yes	183 (97.3%)	193 (98%)	321 (97.3%)		78 (97.5%)	79 (98.7%)	77 (96.2%)
Neural invasion
N0	138 (73.4%)	173 (87.8%)	282 (85.5%)	<0.001	59 (73.8%)	69 (86.2%)	67 (83.8%)	0.101
Yes	50 (26.6%)	24 (12.2%)	48 (14.5%)		21 (26.2%)	11 (13.8%)	13 (16.2%)
LVSI
N0	160 (85.1%)	190 (96.4%)	286 (86.7%)	<0.001	71 (88.8%)	77 (96.2%)	63 (78.8%)	0.003
Yes	28 (14.9%)	7 (3.6%)	44 (13.3%)		9 (11.2%)	3 (3.8%)	17 (21.2%)

Abbreviations: NCRT, neoadjuvant chemoradiotherapy; NCT, neoadjuvant chemotherapy; S+CRT, surgery followed by chemoradiotherapy; LVSI, lymphovascular invasion.

Table 5 Comparison of Pathological Outcomes Between the NCT Group and NCT Group Before and After Propensity Score

Pathological Outcomes	Before Matching			After Matching
	NCT (n=188)	NCRT (n=197)	P value	NCT (n=80)	NCRT (n=80)	P value
Pathological response
Non-pCR	181 (96.3%)	114 (57.9%)	<0.001	78 (97.5%)	52 (65.0%)	<0.001
pCR	7 (3.7%)	83 (42.1%)		2 (2.5%)	28 (35.0%)
TRG
0	12 (6.4%)	102 (51.8%)	<0.001	5 (6.2%)	39 (48.8%)	<0.001
1	17 (9.0%)	26 (13.2%)		4 (5.0%)	8 (10.0%)
2	72 (38.3%)	43 (21.8%)		30 (37.5%)	20 (25.0%)
3	87 (46.3%)	26 (13.2%)		41 (51.2%)	13 (16.2%)
ypT stage
T0/Tis	15 (8.0%)	98 (49.7%)	<0.001	7 (8.8%)	37 (46.2%)	<0.001
T1	9 (4.8%)	13 (6.6%)		1 (1.2%)	3 (3.8%)
T2	37 (19.7%)	32 (16.2%)		14 (17.5%)	11 (13.8%)
T3	119 (63.3%)	50 (25.4%)		53 (66.2%)	27 (33.8%)
T4	8 (4.3%)	4 (2.0%)		5 (6.2%)	2 (2.5%)
ypN stage
N0	73 (38.8%)	155 (78.7%)	<0.001	36 (45.0%)	63 (78.8%)	<0.001
N1	67 (35.6%)	29 (14.7%)		28 (35.0%)	13 (16.2%)
N2	34 (18.1%)	10 (5.1%)		11 (13.8%)	3 (3.8%)
N3	14 (7.4%)	3 (1.5%)		5 (6.2%)	1 (1.2%)

Abbreviations: NCRT, neoadjuvant chemoradiotherapy; NCT, neoadjuvant chemotherapy; pCR, complete pathologic response; TRG, tumor regression grade.

Figure 1 Unmatched Kaplan-Meier survival analysis of DFS (A) and OS (B) among NCT, NCRT and S+CRT groups. Propensity-matched Kaplan-Meier survival analysis of DFS (C) and OS (D) among NCT, NCRT and S+CRT groups.

Abbreviations: OS, overall survival; DFS, disease-free survival; NCRT, neoadjuvant chemoradiotherapy; NCT, neoadjuvant chemotherapy; S+CRT, surgery followed by chemoradiotherapy.

Table 6 Multivariate Analyses of Prognostic Factors Related to OS and DFS Before Propensity Score

Clinical Characteristics	OS		DFS
	HR (CI%)	P value	HR (CI%)	P value
Therapy mode
S+CRT^†	1.58 (1.01, 2.45)	0.043	1.37 (0.95, 1.96)	0.090
NCT^†	0.96 (0.62, 1.49)	0.858	0.80 (0.56, 1.15)	0.226
Covariates
Age (< 60 vs ≥ 60)	1.28 (0.98, 1.66)	0.067	1.08 (0.86, 1.35)	0.510
Sex (M vs F)	0.69 (0.48, 0.99)	0.045	0.82 (0.60, 1.13)	0.221
Smoking (Yes vs No)	1.07 (0.80, 1.43)	0.651	1.06 (0.82, 1.36)	0.661
Alcohol (Yes vs No)	1.09 (0.82, 1.45)	0.542	0.93 (0.73, 1.20)	0.584
Tumor localization (upper vs middle, lower)	0.67 (0.45, 1.01)	0.055	0.65 (0.45, 0.92)	0.017
Surgical approach (OE vs MIE)	1.19 (0.87, 1.64)	0.282	1.08 (0.82, 1.43)	0.569
Surgical types (Sweet vs McKeown)	0.74 (0.54, 1.02)	0.064	0.94 (0.71, 1.23)	0.636
Lymph node dissection counts (< 15 vs ≥ 15)	0.86 (0.64, 1.17)	0.338	0.90 (0.69, 1.18)	0.467
(y)pT stage (T0 vs T1, T2, T3, T4)	3.59 (1.68, 7.68)	0.001	3.04 (1.74, 5.30)	<0.001
(y)pN stage (N0 vs N1, N2, N3)	1.78 (1.34, 2.37)	<0.001	1.99 (1.54, 2.56)	<0.001
Clinical T stage (T1, T2 vs T3, T4)	1.19 (0.86, 1.65)	0.284	1.18 (0.89, 1.58)	0.254
Clinical N stage (N0 vs N +)	1.24 (0.91, 1.68)	0.174	1.04 (0.80, 1.35)	0.780
Neural invasion (Yes vs No)	1.29 (0.94, 1.76)	0.113	1.15 (0.88, 1.52)	0.305
LVSI (Yes vs No)	1.30 (0.91, 1.86)	0.151	1.23 (0.89, 1.68)	0.209

Note: ^†Use NCRT as the reference.

Abbreviations: OS, overall survival; DFS, disease-free survival; HR, hazards ratio; CI, confidence interval; M, male; F, female; NCRT, neoadjuvant chemoradiotherapy; NCT, neoadjuvant chemotherapy; S+CRT, surgery followed by chemoradiotherapy; MIE, minimally invasive esophagectomy; OE, open esophagectomy.

Data Sharing Statement

The supporting data for this study can be procured from the corresponding author, provided a reasonable request is made.