Table 2.

Univariate and multivariable analysis on patient cohort in TMA analysis.

	Univariate analysis
	Logistic regression—CR		Cox regression—OS
	OR (95% CI)	p-value	HR (95% CI)	p-value
Age (year)	1.01 (0.96–1.06)	0.71	1.03 (1.00–1.06)	0.05
Gender (Male vs. female)	0.48 (0.17–1.37)	0.17	1.05 (0.57–1.95)	0.88
BMI pre-treatment	1.05 (0.96–1.15)	0.27	0.94 (0.89–0.99)	0.03
ECOG (0/1 vs. 2/3)	1.64 (0.33–8.23)	0.55	3.3 (1.90–5.73)	<0.001
Smoking (Never vs. past/current)	0.33 (0.10–1.08)	0.07	1.47 (0.82–2.64)	0.2
Diabetes (No vs. Yes)	0.67 (0.24–1.87)	0.44	1.04 (0.59–1.84)	0.89
LVI (No vs. Yes)	0.51 (0.19–1.41)	0.20	1.81 (1.05–3.12)	0.03
Tumor Stage (T2 vs. T3)	0.66 (0.15–2.88)	0.58	3.36 (1.74–6.47)	<0.001
PMN/CD8 tumor (Low vs. High)	0.21 (0.05–0.75)	0.02	1.89 (1.00–3.53)	0.04
NETs post-RT (No vs. Yes)	NA		2.64 (1.23–5.62)	0.01

	Multivariable analysis
	Logistic regression—CR		Cox regression—OS
	OR (95% CI)	p-value	HR (95% CI)	p-value
LVI (No vs. Yes)			2.22 (1.15–4.28)	0.02
PMN/CD8 in tumor (Low vs. High)	0.18 (0.04–0.72)	0.02
NETs post-RT (No vs. Yes)	NA		3.75 (1.67–8.44)	<0.001

Multivariable logistic regression was performed to assess the role of each covariates on a complete response post-RT. In order to estimate the impact of each variable on overall survival a Cox model was performed. Log linearity hypothesis were verified for quantitative variables and proportional hazard assumptions were verified for all variables. Final multivariate model was created using a stepwise selection method. Statistical significance was set at 5%.

CR complete response, OS overall survival, OR odds ratio, HR hazard ratio, BMI body-mass index, NA Non applicable.