INTRODUCTION

Recent data suggest patients with early-onset rectal cancer (EORC) receive neoadjuvant radiation above recommended doses without oncologic benefit. The use of excessive radiation may lead to worse outcomes and patient harm. We sought to evaluate predictors of aggressive neoadjuvant radiation (A-XRT) use in EORC patients and compare this to late-onset rectal cancer (LORC) patients.

METHODS

The National Cancer Database from 2004 to 2014 was queried for rectal adenocarcinoma patients undergoing surgical resection. Patients with stage 0 or IV disease, positive margins, and incomplete data were excluded. Standard neoadjuvant radiation (S-XRT) was based upon NCCN guidelines: 25-50.4 Gray for stage II/III patients and none for stage I. Excess radiation was considered A-XRT. Patients diagnosed at age < 50 years were labeled EORC; those ≥ 50 years were LORC. Categorical data were analyzed with chi-square test. Logistic regression was used to analyze clinicodemographic associations with A-XRT.

RESULTS

45,403 patients were included: 7999 (17.6%) EORC and 37,404 (82.4%) LORC. Multivariable logistic regression demonstrated that A-XRT use among stage I patient was associated with male gender, age under 50, urban location, mucinous histology, and poor tumor differentiation. Among stage II and III patients, A-XRT use was associated with male gender, age under 50, higher education and income, and urban location. Cox hazards did not demonstrate a significant association of A-XRT use with survival.

CONCLUSION

Our data reaffirm that EORC patients more frequently receive A-XRT and that use is based on demographic features independent of tumor characteristics. Reasons for A-XRT, particularly in EORC patients, should be clarified to promote adherence to guidelines and minimize patient harm.