Establishment and validation of a prognostic risk classification for patients with stage T1-3N0M0 esophageal squamous cell carcinoma

doi:10.21203/rs.3.rs-414101/v1

Download PDF

Research

Establishment and validation of a prognostic risk classification for patients with stage T1-3N0M0 esophageal squamous cell carcinoma

https://doi.org/10.21203/rs.3.rs-414101/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: No existing methods can accurately predict postoperative prognosis in patients with stage T1-3N0M0 esophageal squamous cell carcinoma (ESCC). We explored the postoperative prognosis of such patients using a new prognostic risk classification system (Esorisk).

Methods: Patients with stage T1-3N0M0 ESCC from two cancer centers—Sun Yat-sen University Cancer Center (training cohort: N=819) and Henan Cancer Hospital (external validation cohort: N=177)—who underwent esophagectomy between 1995 and 2015 were included. We integrated significant risk factors for death events into multivariable logistic regression models and applied them to the training cohort to build Esorisk. The parsimonious aggregate Esorisk score was calculated for each patient; the training set was divided into three prognostic risk classes according to the 33rd and 66th Esorisk score percentiles. Esorisk’s association with cancer-specific survival (CSS) was assessed using Cox proportional hazard regression analyses.

Results: The Esorisk model was: [10 + 0.023 × age + 0.517 × drinking history – 0.012 × hemoglobin – 0.042 × albumin – 0.032 × lymph nodes removed]. Patients were grouped into three classes—Class A (5.14-7.26, low risk), Class B (7.27-7.70, middle risk), and Class C (7.71-9.29, high risk). In the training group, the 5-year CSS decreased across categories (A: 63%, B: 52%, C: 30%; log-rank P<0.001). Similar findings were observed in the validation group (A: 81%, B: 69%, C: 48%; log-rank P<0.001). Esorisk stratified the CSS in patients from two cohorts with stage pT1 (P=0.0029), pT2 (P<0.001), or pT3 (P<0.001) disease. Cox proportional hazard regression analyses showed that the Esorisk aggregate score remained significantly associated with CSS in the training cohort (P<0.001) and validation cohort (P=0.001) after adjusting for pT stage and tumor differentiation.

Conclusion: We established and validated a novel prognostic risk classification that can predict the CSS for patients with T1-3N0M0 ESCC. This may help clinicians identify patient subgroups with poor prognosis.

Oncology

General Surgery

esophageal squamous cell carcinoma

prognostic model

survival

stage T1-3N0M0

surgery

Table 1: The associations of clinicopathological characteristics between training cohort (SYSUCC) and external validation cohort (HNCH).

	*All patients* *(N = 996)*	*Training Cohort (SYSUCC, N=819)*	*Validation Cohort (HNCH, N=177)*
*Variables*	*No. of patients/ mean ± std*
Sex
Male	680 (68.3%)	577 (70.5%)	103 (58.2%)
Female	316 (31.7%)	242 (29.5%)	74 (41.8%)
Smoking history
No	416 (41.8%)	318 (38.8%)	98 (55.4%)
Yes	580 (58.2%)	501 (51.2%)	79 (44.6%)
Drinking history
No	703 (70.6%)	590 (72.0%)	113 (63.8%)
Yes	293 (29.4%)	229 (28.0%)	64 (36.2%)
Tumor differentiation
Grade I	300 (30.1%)	226 (27.6%)	74 (41.8%)
Grade II	474 (47.6%)	388 (47.4%)	86 (48.6%)
Grade III	222 (22.3%)	205 (25.0%)	17 (9.6%)
T stage
T1	187 (18.8%)	125 (15.3%)	62 (35.0%)
T2	287 (28.8%)	244 (29.8%)	43 (24.3%)
T3	522 (52.4%)	450 (54.9%)	72 (40.7%)
Surgical approach
Sweet	NA	538 (65.7%)	NA
Ivor-Lewis	NA	23 (2.8%)	NA
McKeown	NA	129 (23.1%)	NA
Other	NA	69 (8.4%)	NA
Age (year)	59.3 ± 9.08	58.4 ± 9.24	63.5 ± 6.92
Hemoglobin (g/L)	135.6 ± 14.59	136.6 ± 14.39	130.6 ± 14.54
Albumin (g/L)	43.20 ± 3.89	43.16 ± 3.94	43.06 ±3.69
Lymph nodes	18.15 ± 12.50	18.3 ± 13.0	17.6 ± 9.2
Transthoracic laterality
Left	648 (65.1%)	538 (65.7%)	110 (62.1%)
Right	279 (28.0%)	212 (25.9%)	67 (37.9%)
Other	69 (6.9%)	69 (8.4%)	0 (0%)

SYSUCC, Sun Yat-sen University Cancer Center; HNCH, Henan Cancer Hospital; std, Standard deviations.

Table 2. The results of multivariable Logistic regression analysis in ESCC patients with stage T1-T3N0M0 (dependent variable: death events).

Covariate	Coefficient estimate	Odds ratio	95% Confidence interval	P-value
Age	0.023	1.023	1.007-1.040	0.006
Drinking history
No vs. yes	0.517	1.676	1.194-2.353	0.003
Hemoglobin (g/L)	-0.012	0.988	0.977-0.999	0.038
Albumin (g/L)	-0.042	0.958	0.922-0.996	0.034
Sex
Female vs. male	0.163	1.177	0.817-1.695	0.382
Surgical approach				0.506
Sweet		1	reference
Ivor-Lewis	0.604	1.830	0.756-4.425	0.180
McKeown	-0.022	0.979	0.657-1.457	0.915
Other	-0.172	0.842	0.489-1.449	0.535
Lymph nodes	-0.032	0.969	0.955-0.982	<0.001

ESCC, esophageal squamous cell carcinoma.

Age, hemoglobin, albumin and lymph nodes were numeric variables.

The method was “Enter selection”.

Table 3. Univariable and multivariable Cox regression of stage T1-T3N0M0 ESCC patients in the Sun Yat-sen University Cancer Center.

	Univariable Cox regression		Multivariable Cox regression
Covariate	P-value	HR (95% CI)	P-value	HR (95% CI)
Sex
Female		1 (reference)		1 (reference)
Male	0.899	1.016 (0.799-1.292)	0.197	1.177 (0.919-1.507)
Age (continuous)	0.008	1.016 (1.004-1.028)	0.972	1.000 (0.987-1.013)
Smoking history
No		1 (reference)
Yes	0.084	1.220 (0.973-1.529)
pT stage	0.012		0.001
T1		1 (reference)		1 (reference)
T2	0.05	1.457 (0.999-2.125)	0.04	1.490 (1.019-2.177)
T3	0.003	1.692 (1.191-2.405)	<0.001	1.935 (1.354-2.766)
Tumor differentiation	0.006		<0.001
Grade I		1 (reference)		1 (reference)
Grade II	0.947	1.009 (0.776-1.312)	0.413	1.117 (0.857-1.457)
Grade III	0.007	1.479 (1.111-1.969)	<0.001	1.763 (1.314-2.366)
Esorisk class	<0.001		<0.001
Class A		1 (reference)		1 (reference)
Class B	0.010	1.477 (1.100-1.984)	0.007	1.544 (1.129-2.111)
Class C	<0.001	2.316 (1.762-3.044)	<0.001	2.416 (1.748-3.339)
Surgical approach	0.001		0.014
Sweet		1 (reference)		1 (reference)
Ivor-Lewis	0.041	1.793 (1.025-3.138)	0.014	2.152 (1.68-3.964)
McKeown	0.027	0.727 (0.548-0.965)	NA	NA
Other	0.014	1.615 (1.103-2.366)	NA	NA
Transthoracic laterality	0.007		0.121
Left		1 (reference)		1 (reference)
Right	0.120	0.812 (0.625-1.056)	0.802	0.962 (0.714-1.298)
Other	0.014	1.613 (1.101-2.363)	0.046	1.485 (1.007-2.191)

ESCC, esophageal squamous cell carcinoma; CI, confidence interval; The method was “Enter selection”.

Table 4. Multivariable Cox regression of stage T1-T3N0M0 ESCC patients in the Henan Cancer Hospital.

	Multivariable Cox regression
Covariate	P-value	HR (95% CI)
Sex
Female		1 (reference)
Male	0.714	1.113 (0.627-1.977)
Age (continuous)	0.849	1.004 (0.961-1.050)
pT stage	<0.001
T1		1 (reference)
T2	0.343	1.580 (0.614-4.064)
T3	<0.001	4.411 (2.068-9.412)
Tumor differentiation	0.885
Grade I		1 (reference)
Grade II	0.873	1.048 (0.589-1.863)
Grade III	0.686	0.798 (0.267-2.387)
Esorisk class	0.001
Class A		1 (reference)
Class B	0.699	0.802 (0.262-2.455)
Class C	0.019	3.113 (1.225-7.914)
Transthoracic laterality
Left		1 (reference)
Right	0.343	1.329 (0.739-2.391)

ESCC, esophageal squamous cell carcinoma; CI, confidence interval; The method was “Enter selection”.

Download PDF

Version 1

posted

You are reading this latest preprint version

Establishment and validation of a prognostic risk classification for patients with stage T1-3N0M0 esophageal squamous cell carcinoma

Status:

Version 1

Abstract

Figures

Full Text

Tables

Status:

Version 1