Induction therapy does not complicate minimally invasive esophagectomy

Major finding: There were no significant differences in postoperative deaths or adverse outcomes among patients who underwent induction therapy and surgery for esophageal adenocarcinoma or surgery alone.

Data source: Comparison study of 375 patients with propensity-matched scoring.

Disclosures: The study was internally funded. Dr. Nason reported having no financial disclosures.


AT SSO 2014

PHOENIX, ARIZ. – Induction therapy does not appear to significantly increase the risk of postoperative complications in patients who undergo minimally invasive esophagectomy for esophageal adenocarcinoma, according to results of a study reported at the annual Society of Surgical Oncology Cancer Symposium.

"After balancing pretreatment variables that can potentially influence treatment decisions, we found that induction therapy does not significantly impact on perioperative outcomes compared with patients who are treated with minimally invasive esophagectomy as primary therapy," said Dr. Katie Sue Nason, of the department of cardiothoracic surgery at the University of Pittsburgh.

That conclusion comes from a propensity-matched analysis in which patients with similar pretreatment predictor variables were paired in an attempt to reduce comparison biases.

Dr. Katie S. Nason

The investigators found no significant differences in mortality, major adverse events, readmissions, reoperations, or length of stay between 197 patients who received induction therapy and minimally invasive esophagectomy and 178 who had esophagectomy alone.

Although induction chemoradiation therapy may reduce the incidence of local and distant treatment failures in patients with esophageal cancer, it has the potential to increase the risk of postoperative adverse outcomes.

"What’s often not considered is that these factors that influence postoperative outcomes may also be influencing treatment allocation," Dr. Nason said.

In observational studies, for example, there may be larger differences in observed covariates between treatment groups that could lead to biased estimates of treatment effects.

"This could be adjusted for using propensity score matching, where you generate the conditional probability of one individual being treated with a particular treatment approach given multiple pretreatment covariates. By doing propensity score matching, you can then balance these covariates such as age and various comorbid illnesses between the two groups, and perhaps eliminate this treatment allocation bias that impacts on the relationship between the treatment and the postoperative outcomes," Dr. Nason explained.

She and her colleagues applied the technique to an analysis of outcomes from 375 patients with clinical stage II or greater esophageal adenocarcinoma treated with minimally invasive esophagectomy. They assessed tumor variables, comorbidities, treatments, and outcomes, and created propensity matching scores to match surgery-only patients one-on-one with no repeats to a patient who also underwent induction therapy and had a propensity score within 0.05 of that for the surgery-only patient.

Patients without suitable matches were excluded from the data set.

The extensive list of variables included age, smoking status, alcohol use, history of Barrett’s esophagus, myriad comorbidities, cancer location, pretreatment clinical stage, and many others.

Among the 375 patients, the investigators were able to generate propensity-matching scores for 82 pairs for the comparison of treatments and outcome.

They found that there were no significant differences between induction and surgery-only patients in the primary outcome of adverse events within 30 days of surgery, including in-hospital and 30-day mortality, major adverse events or at least 1 postoperative adverse event, readmission within 30 days, reoperation in-hospital or within 30 days, or length of stay greater than 10 days.

Dr. Nason noted that although unmatched patients differed in age, presentation with alarm symptoms, daily alcohol use, clinical stage and comorbid illnesses, the pretreatment variables were all well balanced in the propensity-matched analysis.

Mortality after minimally invasive esophagectomy was 1.8% among all 375 patients, 1.5% among patients who underwent induction, and 2.3% among patients who underwent surgery alone; these differences were not significant.

Major adverse events occurred in 28% of patients overall, 27% of those who received induction, and 30% of those who had surgery alone, also not significant.

The study was internally funded. Dr. Nason reported having no financial disclosures.

Next Article:

   Comments ()