Pralatrexate and Oxaliplatin in Treating Patients With Unresectable or Metastatic Esophageal, Stomach, or Gastroesophageal Junction Cancer

• Overall response rate to combination pralatrexate and oxaliplatin as assessed by Response Evaluation Criteria in Solid Tumors version 1.1 [ Time Frame: Up to 5 years ]
  Objective responses will be confirmed 4 weeks after first documentation of response.
  
  

• Incidence of toxicity as assessed by National Cancer Institute Common Terminology Criteria for Adverse Events version 4.0 [ Time Frame: Up to 30 days after the last dose of study drug(s) ]
  
• Overall survival (OS) [ Time Frame: From the date of study enrollment to the time of death from any cause, assessed up to 5 years ]
  Estimated using the Kaplan-Meier method and proportional hazards models. Continuous variables will be summarized with commonly used statistics (mean, standard deviation, median, etc.), with sub-group associations tested using the Wilcoxon rank sum test. Categorical variables will be summarized in contingency tables, with associations of interest assessed using Fisher's exact test.
  
  
• Time to progression (TTP) [ Time Frame: From the date of study enrollment to the first observation of progressive disease, assessed up to 5 years ]
  Estimated using the Kaplan-Meier method and proportional hazards models. Continuous variables will be summarized with commonly used statistics (mean, standard deviation, median, etc.), with sub-group associations tested using the Wilcoxon rank sum test. Categorical variables will be summarized in contingency tables, with associations of interest assessed using Fisher's exact test.
  
  

• Dichotomous genomic events (i.e., homozygote/heterozygote, or presence/absence of polymorphisms, etc.) [ Time Frame: Baseline or any time during study ]
  Correlations of the haplotype-tagged single nucleotide polymorphism (SNPs) with response, OS, and TTP will be explored by multivariate logistic and Cox proportional hazards models, to explore the impact of these germline SNPs on binary and time to event outcomes. These exploratory comparisons will be carried out on the entire cohort, as well as within different patient subgroups such as responders, and progression-free and stable patients if feasible.
  
  
• MicroRNA expression profiles [ Time Frame: Baseline ]
  Initial studies will be conducted to test and optimize individual miRNAs for maximum sensitivity and specificity to predict chemosensitivity, inferring differences in the distributions of biomarkers with stringent statistical multiple testing error control procedures. A 2-sample two-ways test of mean differences in mir log expression by discrete treatment outcome (toxicity or efficacy) will be performed, with a Wilcoxon rank sum test.
  
  

PRIMARY OBJECTIVES:

I. To determine the overall response rate in patients with advanced esophago-gastric cancer (EGC) to combination pralatrexate and oxaliplatin.

SECONDARY OBJECTIVES:

I. To examine the toxicity and tolerability of this regimen. II. To determine the time-to-progression and overall survival using this regimen.

III. To examine whether functionally relevant polymorphisms of genes of the folate metabolism pathway correlate with efficacy and toxicity of pralatrexate.

IV. To examine whether response to pralatrexate can be predicted by micro-ribonucleic acid (microRNA) expression profiling of the epithelial component of the tumor.

OUTLINE:

Patients receive pralatrexate intravenously (IV) over 3-5 minutes and oxaliplatin IV over 2 hours on day 1. Courses repeat every 2 weeks in the absence of disease progression or unacceptable toxicity. Oxaliplatin will be discontinued after 12 courses.

After completion of study treatment, patients are followed up for 30 days and then periodically thereafter for up to 5 years.