Background: In published studies, complete response (CR) to chemoradiation occurs in only 25-30% of patients with locally advanced esophageal cancer. Liquid nitrogen spray cryotherapy (LNSC) is postulated to stimulate an anti-tumor immune response. In a preliminary study, the investigators documented CR rate of 56% with a single session of LNSC administered prior to chemoradiation. Before proceeding with larger trials to corroborate these findings, the maximally tolerated dose (MTD) of neoadjuvant LNSC must be determined. The aims of this study are: (1) To determine safety and MTD of LNSC during neoadjuvant chemoradiation in locally advanced esophageal cancer. (2) To assess whether LNSC results in immunogenic cell death. (3) To assess changes in tumor micro-environment with LNSC.

Methods: Eligible adult patients with locally advanced esophageal cancer will receive LNSC at the following dose frequencies: Patient 1, 2, and 3: 2 sessions of LNSC prior to chemoradiation (chemoXRT); Patients 4, 5, and 6: 2 sessions LNSC prior to chemoXRT, then 1 session during week 4 of chemoXRT; Patients 7, 8, and 9: 2 sessions LNSC prior to chemoXRT, then 1 session during week 2 and 1 session during week 4 of chemoXRT. If no dose limiting toxicity (DLT) occurs, the investigators will enroll an additional 3 patients to confirm MTD. The investigators will contact patients at 48-hours and 1-week post-procedure to evaluate for adverse events (AEs) and DLTs, and assess for improvements in dysphagia and quality of life (QOL) using the Mellow-Pinkas and EORTC QLQ-OES18 instruments respectively. The investigators will obtain peripheral blood for ELISA and biopsies from the tumor to assess tumor-infiltrating lymphocytes (TILs) and T cell subtypes before the 1st session of LNSC, before the 2nd session of LNSC, and after chemoradiation is completed.

Expected results: (1) Dose limiting toxicity (DLT) does not occur when patients received 2 session of LNSC prior to chemoXRT, and 2 sessions during chemoXRT (2) LNSC results in immunogenic cell death, as assessed by increased levels of HMGB1 in serum, and calreticulin in biopsy specimens (CRT) (3) LNSC is associated with increased T cell infiltration and activation (increased TILs, CD8+, CD3+ T cells, and granzyme B), and decrease in regulatory T cells (CD45R0, FOXP3).

LNSC will be administered as follows: up to three cycles of 20 to 40 seconds each directed at the bulkiest aspect of the tumor. During each cycle, a surface area of approximately 2 to 3 cm2 is targeted. Up to three sites are treated during each session. Tissue thawing is verified between cycles by waiting at least 60 seconds before re-treatment, and by observing a return to baseline color of the tissue and complete disappearance of ice crystals after reperfusion. After the procedure is completed, patients will be monitored in the recovery unit for at least 30 minutes. Patients will be discharged after monitoring if they meet the endoscopy unit's standard discharge criteria.

Cryotherapy dose frequency intervals

1. Two LNSC sessions (patient 1, 2, and 3): 2 sessions of LNSC administered at 2-week intervals, followed by concurrent chemoradiation for 5 weeks.

2. Three LNSC sessions (patients 4, 5, and 6): 2 sessions of LNSC administered at 2-week intervals, followed by concurrent chemoradiation for 5 weeks. During week 4 of chemoradiation, they will receive a session of LNSC.

3. Four LNSC sessions (patients 7, 8, and 9): 2 sessions of LNSC administered at 2-week intervals, followed by concurrent chemoradiation for 5 weeks. During week 2 and week 4 of chemoradiation, they will receive LNSC (i.e. - 2 sessions during chemoradiation).

   • No DLTs: If no patient develops a DLT, then the investigators will enroll an additional 3 patients to receive 4 LNSC sessions as described earlier, in order to confirm safety.
   • 1 DLT: If 1 of 3 patients develops a DLT, then the investigators will enroll 3 additional patients to receive LNSC at that dose. If any of the additionally enrolled patients develop a DLT, the investigators will enroll 3 patients at the previous dose, to confirm safety of the lower dose frequency.
   • 2 DLTs: At a given dose frequency interval, if 2 of 3 patients develop a DLT, then the investigators will enroll 3 patients at the previous dose, to confirm safety of the lower dose frequency.

Definition of Adverse Events (AEs) and Dose Limiting Toxicity (DLT) We will define adverse AEs based on the American Society of Gastrointestinal Endoscopy (ASGE) lexicon (Appendix 2). The relationship of the AE to LNSC will be classified using the following 4 categories: definitely related, likely related, unlikely related, and definitely not related. DLT will be defined as any severe or fatal adverse event that is definitely or likely related to the cryotherapy procedure (i.e. - unplanned admission > 10 nights, intensive care unit admission > 1 night, surgery for an adverse event, permanent disability due to an adverse event, or death). A member of the study team will contact patients and review their medical records 48 hours after the procedure, and 1 week after the procedure to assess for mild, moderate, and severe adverse events (AEs). In addition to recording any AEs, the study coordinator will contact a physician investigator if any AEs are reported in order to direct appropriate clinical management. An independent data safety monitoring (DSM) committee will review AEs every 3 months for the duration of the study to determine whether the AE was definitely or likely related to the LNSC procedure. Additionally, if a serious adverse event (SAE) is reported, then the DSM will meet within one week to determine if the SAE represents a DLT.

Tissue morphology, flow cytometry, and immunohistochemistry (IHC) The investigators will obtain peripheral blood and 6 mucosal biopsies (2 mm X 2 mm) immediately prior to the first and second session of cryotherapy, and at the end of chemoradiation. The investigators will obtain half the biopsies from the center, and half the biopsies from the periphery of the tumor. If the patient proceeds directly to surgery following chemoradiation without endoscopy, then the investigators will obtain archival tissue from the explant specimen. The investigators will place 2 biopsies in 4% paraformaldehyde (PFA) for TILs analysis (1 center and 1 periphery), 2 biopsies in 4% PFA for IHC analysis (1 center and 1 periphery), and 2 biopsies for exploratory analyses/backup. To assess for evidence of immunogenic cell death, the investigators will perform IHC on biopsy specimens for CRT, and ELISA on peripheral blood for HMBG1. A pathologist blinded to patient demographics, and treatment allocation will grade overall inflammation using standard criteria and as sparse, moderate, or pronounced TILs. To assess the immune infiltrate, the investigators will construct tissue microarray (TMA) blocks and analyze in an automated immunostainer for antibodies against the following antigens: CD3, CD8, CD45R0/FoxP3, and Pan-CK.

Other Data Collection Baseline variables: Baseline variables will include age, gender, ethnicity, tumor histology and TNM stage, location and length of tumor, presence or absence of critical organ invasion (i.e. - aorta, trachea, heart, and vertebral body), ECOG performance status, baseline Mellow-Pinkas dysphagia score, and a validated measure of quality of life in esophageal cancer, the EORTC QLQ-OES18.

Assessment of pathologic/clinical complete response: At the completion of neo-adjuvant chemoradiation, results of the re-staging PET-CT and surgical explant will be abstracted to document presence or absence of a pathologic complete response. As is standard protocol at our institutions, patients who do not undergo surgery will undergo upper endoscopy with biopsies along with PET-CT to assess for the presence or absence of a clinical complete response.

Assessment of palliation efficacy: A member of the study team will contact each patient at 1-week after each LNSC procedure to update the Mellow-Pinkas dysphagia score and EORTC QLQ-OES18. The investigators will also document if there was any need for an esophageal stent, enteral feeding tube, or parenteral nutrition support during chemoradiation.

Inclusion Criteria:

Adult patients with locally advanced esophageal cancer will be eligible to participate in the study. Eligible patients must have a histologically proven diagnosis of esophageal adenocarcinoma, esophageal squamous cell carcinoma, adenosquamous carcinoma, or gastro-esophageal junctional adenocarcinoma. Locally advanced esophageal cancer will be defined as absence of distant metastases on positron emission tomography-computed tomography (PET-CT) and a TNM stage of IIA to IIIC, as defined by endoscopic ultrasound and PET-CT

Exclusion Criteria:

1. Pediatric patients (age less than 18 years)
2. Superficial esophageal cancer
3. Inability to pass the orogastric decompression tube into the stomach
4. Coagulopathy (INR > 2 or platelet count < 50,000/cubic millimeter).
5. Inability to provide informed consent
6. Eastern Cooperative Oncology Group (ECOG) performance status ≥3
7. Pregnancy

Data will be shared in aggregate. No PHI will be shared with other researchers.