Table 3.

Logistic regression analyses of associations between preablation genetic alterations in patients who had persistent BE or recurrent neoplasia compared with controls, adjusting for histology of the tissue sample analyzed (top) and further adjusted for patient age, and BE length (bottom)

Adjusted for histology
	Persistent vs controls Odds ratioa (95% CI)	Recurrent vs controls Odds ratioa (95% CI)
TS	1.21 (0.81–1.81), P = 0.35	1.05 (0.62–1.79), P = 0.85
CNV arms	0.94 (0.82–1.08), P = 0.40	1.03 (0.92–1.15), P = 0.65
Onco amps	1.64 (1.12–2.40), P = 0.01	1.18 (0.67–2.07), P = 0.56
TS dels	2.01 (1.11–3.63), P = 0.02	1.76 (0.91–3.40), P = 0.15
TP53	1.02 (0.36–2.89), P = 0.97	0.83 (0.25–2.72), P = 0.75
APC	1.51 (0.44–5.15), P = 0.51	1.58 (0.41–6.11), P = 0.51
CDK2NA	0.87 (0.24–3.14), P = 0.83	0.73 (0.15–3.45), P = 0.69
RTK	4.52 (1.44–14.15), P = 0.01	1.00 (0.24–4.09), P = 1.00
ERBB2	4.77 (1.05–21.57), P = 0.04	1.40 (0.19–10.15), P = 0.74

Adjusted for histology, age, and BE length
	Persistent vs controls Odds ratioa (95% CI)	Recurrent vs controls Odds ratioa (95% CI)
TS	0.99 (0.61–1.60), P = 0.96	0.90 (0.51–1.60), P = 0.73
CNV arms	0.94 (0.81–1.10), P = 0.47	1.03 (0.92–1.15), P = 0.64
Onco amps	1.35 (0.91–1.99), P = 0.14	1.14 (0.59–2.18), P = 0.70
TS dels	1.94 (0.98–3.84), P = 0.06	1.98 (0.94–4.15), P = 0.07
TP53	0.98 (0.31–3.12), P = 0.98	0.73 (0.21–2.59), P = 0.63
APC	1.77 (0.46–6.84), P = 0.41	1.38 (0.31–6.11), P = 0.67
CDK2NA	0.59 (0.13–2.60), P = 0.48	0.70 (0.15–3.36), P = 0.66
RTK	2.77 (0.77–10.02), P = 0.12	0.98 (0.21–4.66), P = 0.98
ERBB2	1.69 (0.31–9.27), P = 0.55	0.97 (0.10–9.09), P = 0.98

^aShown are odds ratios (95% CI) with bolded estimates indicating statistical significance P < 0.05. Each model includes the individual genomic alteration plus tissue sample histology (e.g., CNV deletions plus histology).

Amps, amplification; BE, Barrett's esophagus; CI, confidence interval; CNV, copy number variant; dels, deletion; Onco, oncogene; RTK, receptor tyrosine kinase; TS, tumor suppressor.