IMPORTANCE High-risk human papillomavirus (HPV) has been associated with Barrett dysplasia and esophageal adenocarcinoma. Nevertheless, the prognostic significance of esophageal tumor HPV status is unknown.
OBJECTIVE To determine the association between HPV infection and related biomarkers in high-grade dysplasia or esophageal adenocarcinoma and survival.
DESIGN, SETTING, AND PARTICIPANTS Retrospective case-control study. The hypothesis was that HPV-associated esophageal tumors would show a favorable prognosis (as in viral-positive head and neck cancers). Pretreatment biopsies were used for HPV DNA determination via polymerase chain reaction, in situ hybridization for E6 and E7 messenger RNA (mRNA), and immunohistochemistry for the proteins p16INK4A and p53. Sequencing of TP53 was also undertaken. The study took place at secondary and tertiary referral centers, with 151 patients assessed for eligibility and 9 excluded. The study period was from December 1, 2002, to November 28, 2017.
MAIN OUTCOMES AND MEASURES Disease-free survival (DFS) and overall survival (OS).
Among 142 patients with high-grade dysplasia or esophageal adenocarcinoma (126 [88.7%] male; mean [SD] age, 66.0 [12.1] years; 142 [100%] white), 37 were HPV positive and 105 were HPV negative. Patients who were HPV positive mostly had high p16INK4A expression, low p53 expression, and wild-type TP53. There were more Tis, T1, and T2 tumors in HPV-positive patients compared with HPV-negative patients (75.7%vs 54.3%; difference, 21.4%; 95%CI, 4.6%-38.2%; P = .02). Mean DFS was superior in the HPV-positive group (40.3 vs 24.1 months; difference, 16.2 months; 95%CI, 5.7-26.8; P = .003) as was OS (43.7 vs 29.8 months; difference, 13.9 months; 95% CI, 3.6-24.3; P = .009). Recurrence or progression was reduced in the HPV-positive cohort (24.3%vs 58.1%; difference, −33.8%; 95%CI, −50.5%to −17.0%; P < .001) as was distant metastasis (8.1%vs 27.6%; difference, −19.5%; 95%CI, −31.8%to −7.2%; P = .02) and death from esophageal adenocarcinoma (13.5%vs 36.2%; difference, −22.7%; 95%CI, −37.0% to −8.3%; P = .01). Positive results for HPV and transcriptionally active virus were both associated with a superior DFS (hazard ratio [HR], 0.33; 95%CI, 0.16-0.67; P = .002 and HR, 0.44; 95%CI, 0.22-0.88; P = .02, respectively [log-rank test]). Positivity for E6 and E7 mRNA, high p16INK4A expression, and low p53 expression were not associated with improved DFS. On multivariate analysis, superior DFS was demonstrated for HPV (HR, 0.39; 95%CI, 0.18-0.85; P = .02), biologically active virus (HR, 0.36; 95%CI, 0.15-0.86; P = .02), E6 and E7 mRNA (HR, 0.36; 95%CI, 0.14-0.96; P = .04), and high p16 expression (HR, 0.49; 95%CI, 0.27-0.89; P = .02).
CONCLUSIONS AND RELEVANCE Barrett high-grade dysplasia and esophageal adenocarcinoma in patients who are positive for HPV are distinct biological entities with a favorable prognosis compared with viral-negative esophageal tumors. Confirmation of these findings in larger cohorts with more advanced disease could present an opportunity for treatment de-escalation in the hope of reducing toxic effects without deleteriously affecting survival.